Desigining and Implementing A Self Learning Expert troubleshooter for PC for Increasing user Satisfaction

Abstract

In today’s world there is tremendous growth in the IT sector and with it’s this huge growth people’s dependence on computers is increasing day by day. As presence of computers is the basic requirement to settle any IT sector then it is always necessary that they always work smoothly without any trouble to the user. So, to help the smooth working of computers i.e. to troubleshoot the PC faults and to diagnose the faults this Self learning Expert PC Troubleshooter (SLEPCT) is developed which is discussed in this thesis work. This will help users to solve their PC related problems by themselves on their desk itself without contacting any technician. And directly it will reduce the work load on the technicians as computer fault diagnosis is automated here. The system is composed of a user interface, a knowledge-base, an inference engine, and an expert interface and a hybrid learning algorithm. Additionally, the system features a mail server that helps admin to get notifications about new queries asked from different users and then helps in responding back to the new queries and sending replies through mails to the users by mails. The learning algorithm embedded in the system enhances the performance of the system as it increases the user satisfaction level by providing the solutions to the faults that are widely accepted by the people. The working of the system is based on the feedback taken from the user itself. So, in all the nominated system frees up human technicians from manually performing laborious and time consuming maintenance tasks.